IMAGINE if you had the recipe for breathing life into inanimate chemicals. This is exactly what a handful of ambitious researchers in labs around the world are planning to do. They aim to discover the essential genes needed to support a simple living cell, stitch together this 鈥渕inimal genome鈥� and then press the on switch.
Building a genome from scratch could be seen as an extension of existing genetic engineering techniques 鈥� albeit one of the most audacious projects ever attempted. But other researchers want to go further and re-engineer life using chemicals not chosen by evolution. Such cells could be nanotechnology factories, or even form the basis of biological computers. It means there could one day be a new form of life on Earth 鈥� and one that is unquestionably the product of intelligent design.
鈥淲e鈥檇 like to rebuild a complete genome, based on knowing the first principles of biology,鈥� says US genomics pioneer Craig Venter, who hit the headlines after announcing his plans to do just that last year. The motivation for Venter and other researchers is to discover fundamental truths about the basis of life. Sure, we know a great deal about biochemistry, and have generated vast stores of information on genes and the proteins they encode. But how does that all fit together? What makes a cell alive?
Advertisement
To find out, scientists are trying to discover the minimum genes needed for a cell to live independently, as these identify the key properties of life. 鈥淭here is something sexy in the phrase 鈥榤inimal gene set鈥�,鈥� says Eugene Koonin, a computational biologist at the National Institutes of Health in Bethesda, Maryland. 鈥淭here is the connotation of discovering the secrets of life.鈥�
Perhaps the first problem is to define what minimal life is. Philosophers and biologists have been tying themselves in knots for years, trying to agree on the meaning of life (New 杏吧原创, 13 June 1998, page 38). In this context, scientists define it as a cell that can live, grow and reproduce independently of other cells. That rules out viruses, which are intracellular parasites. To start with, researchers are trying to work out the minimum gene sets for cells that get all the energy and nutrients they need from their environment. From there they can move on to minimal gene sets for more self-sufficient organisms, ones that make their own energy supply from sunlight, for example.
As well as revealing the secrets of life, minimal genome research should help evolutionary biologists get a picture of the cell that gave rise to all forms of modern cellular life. Dubbed the 鈥渓ast universal common ancestor鈥�, or LUCA, this is the mother of all existing creatures on Earth.
In fact it was the search for LUCA that triggered the idea of the minimal genome, back in the 1960s. Researchers such as biochemist Harold Morowitz, now professor of biology and natural philosophy at George Mason University in Fairfax, Virginia, started to wonder what the simplest living organism was, and whether it would be similar to LUCA. Morowitz suggested the answer might lie with a group of bacteria called mycoplasmas, the smallest known independently living cells.
Freeloaders
Morowitz was on the right track. In 1990, researchers discovered that Mycoplasma genitalium, a parasitic bacterium that infects the human urinary tract, had the smallest genome of any independently living cell. Venter鈥檚 team at The Institute for Genomic Research (TIGR) in Rockville, Maryland, sequenced this species鈥� genome in 1995 and showed that it contained a mere 480 protein-coding genes, compared with some 4100 in the well-studied soil bacterium Bacillus subtilis, say. But it soon became clear that M. genitalium hadn鈥檛 always had such a tiny genome. As a parasitic freeloader, the bacterium had dispensed with numerous genes because it took so many nutrients straight from its host. M. genitalium is no LUCA 鈥� but it is still the subject of intense interest from minimal genome researchers.
It was the dawn of the genomics era that revitalised the search for the minimal genome. For the first time, scientists could compare bacteria鈥檚 DNA to see which genes they shared 鈥� one of the three main approaches to minimal genome research. In 1996, Eugene Koonin at the NIH and his colleague Arcady Mushegian published a comparison of the first two complete bacterial genomes: M. genitalium and Haemophilus influenzae, which causes infections such as pneumonia. Separated by two billion years of evolution, these bacteria seemed ideal candidates. Any genes they had in common must surely represent the core set of genes needed to support cellular life.
The team looked for genes with a similar sequence in both bacteria 鈥� known as 鈥渙rthologue鈥� genes 鈥� and found 256. This was a useful starting point, but it was likely to be only a rough estimate. Over millions of years the two bacteria would have separately evolved different genes to do the same job. These days, with some 100 microbial genomes completely sequenced, it鈥檚 clear that there are very few orthologue genes common to all species.
But that doesn鈥檛 mean comparative genomics has nothing to tell us about the minimal genome. Venter and his team at TIGR have sequenced the genomes of another 12 bacteria that are cousins of the mycoplasmas, and comparing them has helped narrow down the essential gene set, although the researchers have not yet made the details public.
Venter is also following the second route to the minimal genome 鈥� disabling a bacterium鈥檚 genes one by one, to discover which ones are essential for life. In 1999, his team announced they had done this with M. genitalium, the tiny parasite that likes a free lunch. From the starting point of 480 genes, they came up with about 300 essential ones (Science, vol 286, p 2165). Surprisingly, the function of 100 of these was unknown, suggesting there are some gaps in our knowledge of basic cell biology. Venter says: 鈥淚t鈥檚 kind of hard to know all of the first principles when we have no idea what the function is of a third of the genes essential for life in a rich nutrient environment.鈥� TIGR researchers are now hot on the trail of these mystery genes, comparing them with those of other bacteria and stitching them into artificial chromosomes to test their functions.
Other researchers are performing gene-disabling experiments in more species, including the lab workhorse Escherichia coli and B. subtilis. Although these bacteria have much larger genomes than M. genitalium, many of the genetic tools needed are available because of these species鈥� long use in research.
Dusko Ehrlich, a biologist at the National Institute for Agricultural Research in France, and 96 international collaborators embarked on a painstaking experiment to disable almost every one of B. subtilis鈥檚 4100 genes. In a paper published in April (Proceedings of the National Academy of Sciences, vol 100, p 4678), they estimated the minimal gene set to be about 270 genes, fitting well with Venter鈥檚 M. genitalium findings. Unlike Venter, however, the team found that only a dozen of their essential genes had unknown functions.
And Ehrlich and Venter鈥檚 teams are not the only ones close to finding the genes essential for existence. A team led by Andrei Osterman at biotech company Integrated Genomics in Chicago are about to publish a minimal gene set for E. coli. 鈥淲e鈥檙e getting closer to the core machinery that drives life,鈥� says Osterman.
Such research can have more immediate practical benefits too. Fred Blattner at the University of Wisconsin in Milwaukee and Gy枚rgy P贸sfai at the Biological Research Center in Szeged, Hungary, have formed a company, Scarab Genomics, to prune down the E. coli genome to produce a simpler, safer and better-behaved strain for lab use and industrial production. The team compared the genomes of two strains of E. coli 鈥� a lab variety called K12 and the infamous 0157 strain, which causes food poisoning. They looked for genes the two bacteria had in common; the remainder were deemed to be superfluous, unnecessary for growth in the lab dish. So far the team has trimmed the genome by 20 per cent. Many of the superfluous genes gave the bacteria functions that were harmful to humans.
Blattner and his team are now trying to understand what makes bacteria turn even nastier, by comparing E. coli with its deadlier relatives, the typhus bug Salmonella typhi and the plague bacterium Yersinia pestis. After repeated comparison and pruning, the team reckon they will strip down the genome to something akin to the common ancestor of all gut bacteria, and could perhaps retrace evolution鈥檚 steps back further still. 鈥淲e鈥檇 finally get down to something that鈥檚 pretty minimal,鈥� says Blattner.
Instead of stripping genes away from existing bacteria in a top-down approach, the third route to the minimal genome is from the bottom up. With this method, being tried by Venter, among others, you start with nothing, and then add in genes that are essential. This is the best way to prove that the candidate minimal gene set identified by the other two methods really does allow life to exist.
So what is needed for minimal life? First, you need to store information 鈥� in the form of a double helix of DNA 鈥� and to repair this molecule when damaged, translate it into proteins and copy it, to reproduce yourself. About half of the essential genes in B. subtilis do these jobs. Secondly, you need a cell membrane to separate yourself from your environment and give you an identity as an individual life form. Thirdly, you need to carry out chemical reactions to extract energy from nutrients, and to produce structural materials from the basic building blocks of amino acids, sugars and fats. You need a metabolism, in other words. Then there are functions that remain to be discovered, such as those performed by the unknown essential genes.
Once these key functions are known, the next challenge is to synthesise huge lengths of DNA, starting from its building blocks, nucleotides. 鈥淚t鈥檚 extremely difficult,鈥� says Venter. It鈥檚 currently possible to make stretches of DNA that are 5000 base pairs long, but even M. genitalium鈥檚 tiny genome is a hundred-fold bigger. 鈥淢aking something half a million base pairs long is something that has never been attempted 鈥� to my knowledge 鈥� before,鈥� says Venter. His team has developed a new technique for making short pieces of DNA and then gluing them together to make longer ones, although he is keeping the details of the work under wraps for the time being.
The final hurdle is to get the minimal genome to function inside a cell. Venter plans to remove the DNA from an M. genitalium bacterium and insert his artificial genome. Exactly how he is going to persuade the bacterium to swallow 500,000 base pairs of DNA is one problem. Whether the new genome will actually 鈥渂oot up鈥� inside the cell is another. 鈥淐an you take a chromosome from a cell, substitute a new one 鈥� like changing an operating system on a computer?鈥� asks Venter.
Laws of life
Once someone manages to build a functioning minimal genome, what would we do with it? For one thing, such an organism would be the ideal tool for researchers in the hip new field of systems biology. This discipline is the study of how all our detailed knowledge of biochemistry fits together, to discover any overarching laws or principles that govern cells鈥� behaviour. Many biologists argue there must be simple laws governing life, and believe that discovering them will change molecular biology from an experimental, descriptive science into a predictive one. But if you want to foretell the behaviour of a highly complex system like a cell, first you have to intimately understand it. Venter says: 鈥淭rying to understand cellular life with an experimental system where we could add and remove different components and really understand what happens, would be truly wonderful.鈥�
Ever the entrepreneur, Venter also has more ambitious plans for his minimalist creation, and talks of engineering it to produce greener fuels, clean up the environment and reduce global warming. To this end he set up the Institute for Biological Energy Alternatives last year, which has already received $12 million of funding from the US Department of Energy. IBEA scientists are working on an ambitious project to sequence the genomes of hundreds of little-studied marine bacteria in the hope of finding useful genes.
The promise of a minimal cell has also caught the eye of nanotechnologists like George Church at Harvard Medical School. 鈥淲hat are biological systems good at?鈥� asks Church. 鈥淭hey are really good at building things.鈥� This makes them attractive to nanotechnologists who want to produce living nanofactories for such things as DNA memory banks. But to do this, a system whose behaviour is known and predictable is key. So Church is building one from scratch.
Church鈥檚 definition of minimal life is simple: 鈥淵ou want a set of polymers that take in simple monomers to make copies of the polymers,鈥� he says. Then you can work up to more complex systems.
Church is well on the way to achieving his goal. He started by collecting all the enzymes needed to copy DNA and turn it into protein, many of which are commercially available. Church鈥檚 team obtained the genes from E. coli or certain viruses that encode these enzymes, and made tiny 鈥渃hromosomes鈥� to house each gene. The team then mixed all these ingredients together in a glass tube, added some raw materials such as amino acids and nucleotides, and found that the system did indeed copy itself and make proteins. The number of genes needed to do this? A mere 110.
Admittedly, there鈥檚 some way to go from Church鈥檚 system to what we would recognise as a cell 鈥� there鈥檚 no cell membrane for example 鈥� but it鈥檚 an impressive start. 鈥淲e鈥檙e taking baby steps,鈥� says Church.
鈥淭he engineering of biological organisms is maybe the engineering discipline of the century,鈥� says Tom Knight at MIT鈥檚 Artificial Intelligence Laboratory. He also wants to create a simple, understandable and malleable genome, but plans to edit an existing one rather than start from scratch. His team is identifying essential genes in Mesoplasma florum, a relative of the mycoplasmas. They are developing ways to add or remove genes, to make a system that was originally cobbled together by natural selection operate from a more logical, modular design. 鈥淭he program was written four billion years ago,鈥� jokes Knight. 鈥淚t鈥檚 time for a rewrite.鈥�
Church too wants to expand evolution鈥檚 repertoire. He is hoping to re-engineer cells to build proteins with 鈥渁rtificial鈥� amino acids 鈥� ones that don鈥檛 belong to the set of 20 used by nearly all life on Earth. 杏吧原创s are only just beginning to speculate about the enormous potential of brand-new amino acids to revolutionise biochemistry and medicine.
But for many, the prospect of synthetic life raises fears about safety. What would happen if these things escaped into the environment? Could terrorists abuse the knowledge, to build new, deadly bacteria that can overpower our immune defences? It鈥檚 unlikely. For one thing, any minimal organism couldn鈥檛 survive without being mollycoddled in the lab. 鈥淭his thing is going to need to be spoon-fed,鈥� says Church. 鈥淭hat makes it very dependent on humans.鈥� Venter agrees: 鈥淲e would want it so that they would have zero chance of surviving outside the laboratory or the industrial environment.鈥� Deciding how much to make public about these discoveries, however, is a tougher call. 鈥淭he last thing my team wants to do is publish methods that would enable somebody to use the same wonderful technology for evil purposes,鈥� says Venter. 鈥淭hat鈥檚 going to be a real dilemma.鈥�
But discovering and building the minimal genome will revolutionise our understanding of cells and whole organisms. 鈥淥ne just needs to realise that the deep mysteries are not revealed automatically,鈥� says Koonin. 鈥淚t鈥檚 the beginning rather than the end of the road.鈥�